Indicator Lamp Dimming?

[danman]

One way that you could check for a current draw problem in the amp is to plug it into a light bulb current limiter. With all tubes removed, and the amp powered up and off of standby, the bulb in your current limiter should only flash for a brief second after taking the amp off of standby. This brief flash is caused by the filter caps charging up and after a second or two the light bulb should dim out again. If the bulb stays lit then you know that something is drawing current somewhere in the amp. This test won't work correctly with the tubes installed because the heater filaments are drawing power and will cause the bulb to light.

[revolver1]

Hmmmmm so... I've checked all the resistors with a meter and no problems there. I don't have an ESR meter and unfortunately I can't afford to pick one up at the moment so my tests on the capacitors throughout aren't 100% conclusive but I'm confident they're all good.
I checked them all, resistance wise and there's no shorts. Then I charged them up with my meter then swapped the leads round and charged the other side then reversed the leads again to check it had discharged and checked the other side for discharge as well. So they're charging and discharging OK. I then did a capacitance test on them and they're the same as when I installed them.
I know it's not an ESR test but given the tests I did and the infrequency of a cap going bad I'm fairly confident the caps are good.

I also checked the pots and they are fine, I've basically gone through it all.

I've been monitoring the bias and one odd thing the valve that was the hottest one up until just the other day is no longer the hottest one. The other one of the matched pair is now the hottest???

And this is another thing that I'm not sure if it's normal. I got up this morning and biased it up with the wall voltage around 233Vac, I was getting around 469Vdc on the plates. I biased for 65% so I've got around 34.6ma cathode current and the bias voltage is sitting at -42Vdc. So I came back to it later on in the day and I'm getting around 228Vac out of the wall and as expected the plate voltage is lower at around 458Vdc. But here's the bit that gets me the cathode current is now running down probably around 48 to 50% instead of 65% as I had previously set it. In fact this seems to happen all the time, It's only an accurate 65% of the plate voltage at the voltage I set it at, if the wall voltage changes the cathode current doesn't scale up or down at 65%.
For this reason I'm biasing really early in the morning for the highest wall voltages and therefore worst case scenario. But is this normal??? Is it really a case of just get it in the ball park and as long as it sounds good then that's OK? It sounds fine to me and I know I'm nowhere near the danger of red plating them but It kinda seems a bit random.

So it's looking like it may be the valves, unfortunately I can't spring for a fresh set at the moment. But it still seems a little odd again I'm no expert but It's not the kinda thing I would have expected a valve to be responsible for. Guess I can't be truly conclusive until I get a new set sorted. But at the moment unless anyone has any other ideas I can't see it can be much else?

[revolver1]

Sorry danman I've only just got your last post. That sounds like a good test. Just to clarify is that just a regular house hold filament lamp? And is it connected in on the primary side of the p.t?

[danman]

The bias voltage will not track evenly when the plate voltage goes up and down due to fluctuations from the ac line. This is why an amp will sometimes sound different at different times of the day or in different venues. The best scenario is to set it to the highest ac voltage that you normally see throughout the day. The tone may suffer a bit though if the voltage drops later in the day but at least you know your tubes won't over-dissipate.

The bulb for the current limiter is just your standard incandescent light bulb wired in series with the ac hot line. The bulb will protect your amp in the event of a short by limiting the amount of current that can flow. I always use mine when I fire up a new build to check for problems without having to worry about damaging anything in the amp. If the bulb only lights for a second or two and then goes dim, I know everything is wired properly and I should be safe to install the tubes. With the tubes installed the bulb will always glow because of the current that the heater filament circuit draws. If you are concerned about your amp, try removing the tubes and plugging into the limiter to check it. I will normally use a 60 or 75watt bulb in mine but here in the US they are becoming harder and harder to find. I stocked up on quite a few years ago when everything started to go to fluorescent bulbs. Check over in the "Technical Discussion" section at "Ampgarage.com". They have a sticky near the top with instructions for building one.

[revolver1]

Thanks, that explains alot. The biasing thing has been bothering me for a while now, but that all makes sense.

The lamp test sounds good, so I could confirm if its the amp or the valves. The only thing I'm a bit unsure of is at idle there's no problem. It only happens when the gains up around 5 and there's signal going in the front end.

Is it ok to turn the volume up and put signal into the amp with the valves out and wouldn't the amp draw current anyway with signal going in to it?

[danman]

Unfortunately without the tubes installed, the power supply will not be under any load. If there is no load placed upon the supply, then any possible issues will not show themselves. The circuit will also not be able to pass any audio signal without them. I would suggest trying a new set of power tubes, and if the problem still occurs then it's possible that something may be up with the PT. It is very odd that this is occurring on the primary side though.

[danman]

I forgot to ask earlier if the filter caps have been changed since you began to notice this problem with the lamp? I saw in one of your earlier post that the lamp dimmed when you took the amp out of standby. It makes me think that you have a cap that may be pulling excessive current during it's charging cycle or that the PT may have an issue and is not supplying its full rated current while under load.

[revolver1]

Yes, I changed all the filter caps from F&T to ARS almost a year ago now. I was working alot at the time so the amp never got turned on after I chainged them, it sat for about 6 or 7 months. Then when I came to use it again I was using it 3 or 4 times a week for about 3 months before it started playing up.

I did notice something a bit odd with the pt the other day and I honestly can't remember if it was like that when I installed it. If you look at it from the end of the amp not the front, the laminates are leaning to the front and not sitting entirely square.
As th amp was breaking in you could see a little bit of what appears to be melted varnish where the pt rests on the chassis. Almost like it melted with the heat as the pt was breaking in.
This may sound a bit strainge but this is old English houses for ya. The floor where the amp sits in my house has a bit of a slope to it so the amp is on a bit of an angle.
It's almost as if the pt has heated up and made the varnish soft then with all the pt buzz the laminates have slipped foward like the leaning tower of piza.
But I kinda thought it would have shorted out if there was something bad going on on the primary side. Mayby I need an insulation resistance test but I'm not to sure if thats a good idea with a trans former. I'd have to check its safe first.

[danman]

Pt's can get pretty warm if the amp is pushed hard for long periods of time, but I believe that the varnish is made to handle those temps without breaking down. The laminate stack should be held together pretty tightly by the four bolts but I suppose anything is possible.

[revolver1]

Back again, this ones sending me round the twist... I thoroughly searched the web for articles on trouble shooting amps and the process to follow, read a whole lot of great info. I learnt a lot of good stuff in the process but nothing seemes to fit the symptoms 100%.
So I got a full set of valves installed and biased up and much to my disappointment it wasn't that. I found the 100k tone stack resistor was up at 140k so I pulled it and replaced it but It wasn't that.

Then I picked up an ESR meter and went through it all. I have a new set of sozo's to use as a reference, I found one mustard that was 10 ohms higher than the rest so I replaced it just to be sure and It wasn't that. The power caps were slightly higher than expected capacitance wise at 88uf but as they are two 50/50uf cans in series parallel I assume the higher capacitance is from the secondary P.T connection.

I noticed It dims lightly when I first start playing and then dims more heavily as the amp warms up after the first couple of mins. There's four carbon comps in the amp so just to be sure it wasn't one of them going madly out of spec when everything heats up I replaced all of them and it wasn't that either. I'm considering swapping them out for carbon film any way, not totally sure yet but this isn't the time for that anyway.

As I don't have any other amps to reference to, can any one tell me if there primary P.T voltage dips and by how much when hitting a chord. Or does it remain constant and stable?

The ESR meter can't read the two tone stack caps or the P.I cap, I'm 99% sure they're all good but today I'm going to patch a second cap in to each position one at a time and see if I get any joy.
I'm considering pulling the valves and meggering some of the sections of wiring like the heaters just to see if its an insulation problem but nothing looks out of place or damaged in any way. There's not a lot I can check without damaging components anyway.

I'm stumped, any ideas??? I don't want to get into randomly swapping out any more parts. I'm just wasting good components and its not really a good system of fault finding.

[revolver1]

So I've replaced all the big electrolytic cans. One of the power cans was reading a little low capacitance wise, still in spec but I replaced them all anyway. The P.I replacement can was a little higher in capacitance as well, its tightened it all up a little in a good way sounds great but still got the problem.

Its odd because it starts out with barely a flicker but after five minutes or so then it really starts going out. I think its marginally better with the new cans but not 100% fixed.
I replaced the 47k neg feed back and the 2.7k cathode carbon comp resistors with carbon film. Then I replaced the 820ohm on V1a and the 33k carbon comp with new ones and although it hasn't fixed it, it definitely sounds better its got a bit more punch but it hasn't totally lost that carbon comp thang.

I got hold of a 500pf 2kv and a 600pf 5kv silver mica cap and put them in the tone stack, again that didn't fix anything but they sound better than the 500pf 500v mica caps that were in there before.
I got a new indicator lamp that hooks into the heaters rather then the 100v tap. Apparently the neon has a flash point of 95v anything less than that is not enough to light it up. So since there was 98v on the 100v tap and when I hit a chord it dipped to 95v ish I can understand why its going out. The thing I don't understand is why it has been fine for 3 years and only starts playing up now.
Maybe the wall voltage has lowered slightly. Is it possible that as the P.T matures it develops more sag??? I am starting to think one of the trannies may be the culprit.

I was getting 3.14v on the heaters of V7 and the new lamp seems fine on there. There is an ever so slight dip on it when I hit a chord but I probably only notice it because I'm aware of it. I still need to try it with my boost and fuzz pedals but I'll have to wait for the neighbours to go out for that.

One good thing to come from it with the component changes the amp sounds much better when I wind it up then it did before so now I use the pedals less and the amp more it really grinds.

It's just such an odd thing. I'd be very interested to know how much voltage sag other guys are seeing on there primarys. If any one could find out next time they've got their amp on the bench I'd be very great full.

[danman]

Glad to hear that the new light is working out better. If I get a chance one day I'll try to measure the sag on mine. I know that at full tilt there will be a fair amount of voltage drop...even more so with a tube rectifier. This is what creates the compression and sag when they are played at full volume.